Skip to main content Skip to navigation

Professor David Firth (Emeritus Professor)

David Firth came to Warwick in 2003 from Oxford, where he was Professor of Social Statistics. Before that he held academic positions at Imperial College London, at the University of Texas at Austin, and at the University of Southampton. He works on general statistical theory and methods, and applications in several areas of science, social science and public life. From 2023 to 2025 he is a Leverhulme Emeritus Fellow.

David is a former Editor of the Journal of the Royal Statistical Society (Series B, Statistical Methodology), and has served as Associate Editor of several other prominent journals including Biometrika and Biostatistics. In 2008 he was elected a Fellow of the British Academy. In 2012 he was awarded the Guy Medal in Silver.

Consultancy projects have included:

  • election-night forecasting for the BBC, where David developed a novel method for exit polling which has now been used by the major UK broadcasters at the last five UK general elections
  • working with the Social Disadvantage Research Centre in Oxford he developed a new method to combine disparate dimensions of local deprivation data into an overall Index of Multiple Deprivation, the method that has now been in use across all four nations of the UK since 2000.

Current research topics include:

  • inference and computation for generalized nonlinear models (with Heather Turner and Ioannis Kosmidis);
  • inference and computation for complex random-effects models (with Cristiano Varin);
  • models of competition and the analysis of pair-comparison data (with Manuela Cattelan, Cristiano Varin, Heather Turner, David Selby and Ian Hamilton);
  • compositional data analysis (with Fiona Sammut and Zeeshan Ali);
  • penalized likelihood methods, especially for modelling discrete data (with Ioannis Kosmidis).

Selected publications since 2003

For complete list of publications, see David's CV (PDF) and/or Google Scholar, Publons, ORCID, Depsy profiles.

Varin, C and Firth, D (2024). Ridge regression for paired comparisons: A tractable new approach, with application to Premier League football. arxiv:2406.09597

Hamilton, I, Tawn, N and Firth, D (2023). The many routes to the ubiquitous Bradley-Terry model. arXiv:2312.13619

Firth, D and Sammut, F (2023). Analysis of composition on the original scale of measurement. arXiv:2312.10548

Firth, D (2023). Sir David Cox: A Few Memories. Harvard Data Science Review, 5(2). https://doi.org/10.1162/99608f92.f27cc65f

Firth, D, Reid, N, Mayo, D and Battey, H (2022). Remembering Sir David Cox, 1924–2022. Significance 19(2), 30–37. https://doi.org/10.1111/1740-9713.01632

Turner, H L, Batchelor, A D and Firth, D (2022). Nonlinear discrete-time hazard models for women's entry into marriage. Statistical Modelling 22, 107–126. https://doi.org/10.1177/1471082X211062651

Kosmidis, I and Firth, D (2021). Jeffreys-prior penalty, finiteness and shrinkage in binomial-response generalized linear models. Biometrika 108, 71–82. https://doi.org/10.1093/biomet/asaa052

Turner, HL, van Etten, J, Firth, D and Kosmidis, I (2020). Modelling rankings in R: The PlackettLuce package. Computational Statistics 35, 1027–1057. https://doi.org/10.1007/s00180-020-00959-3

Firth, D, Kosmidis, I and Turner, HL (2019). Davidson-Luce model for multi-item choice with ties. http://arxiv.org/abs/1909.07123

Varin, C, Cattelan, M and Firth, D (2016). Statistical modelling of citation exchange between statistics journals (with discussion). Journal of the Royal Statistical Society A 179, 1-63. This paper appeared originally at http://arxiv.org/abs/1312.1794. The zip archive of supplementary material (0.4MB) contains documented data and R-code for the analyses. The paper was read at the Royal Statistical Society on 13 May 2015, ahead of publication (with discussion) in J Roy Stat Soc A.

Cattelan, M, Varin, C and Firth, D (2013). Dynamic Bradley-Terry modelling of sports tournaments. Journal of the Royal Statistical Society C 62, 135–150. The zip archive of supplementary material contains the data and R code used for the examples in the paper.

Turner, H and Firth, D (2012). Bradley-Terry Models in R: The BradleyTerry2 Package. Journal of Statistical Software 48(9), 1–21.

Firth, D (2011). On improved estimation for importance sampling. Brazilian Journal of Probability and Statistics 25, 437–443. (PDF download PDF file)

Kosmidis, I and Firth, D (2011). Multinomial logit bias reduction via the Poisson log-linear model. Biometrika 98, 755–759. (Preprint available as CRiSM research report 10-18)

Varin, C, Reid, N and Firth, D (2011). An overview of composite likelihood methods. Statistica Sinica 21, 5–42. (Preprint available as CRiSM research report 09-45)

Kuha, J and Firth, D (2011). On the index of dissimilarity for lack of fit in log-linear and log-multiplicative models. Computational Statistics and Data Analysis 55, 375–388.

Kosmidis, I and Firth, D (2010). A generic algorithm for reducing bias in parametric estimation. Electronic Journal of Statistics 4, 1097–1112

Kosmidis, I and Firth, D (2009). Bias reduction in exponential family nonlinear models. Biometrika 96, 793–804.

Curtice, J and Firth, D (2008). Exit polling in a cold climate: The BBC/ITV experience in Britain in 2005 (with discussion). Journal of the Royal Statistical Society A, 171, 509–539.

Sergeant, J C and Firth, D (2006). Relative index of inequality: Definition, estimation and inference. Biostatistics 7, 213–224.

Whiting, M J, Stuart-Fox, D M, O'Connor, D, Firth, D, Bennett, N C and Blomberg, S P (2006). Ultraviolet signals ultra-aggression in a lizard. Animal Behaviour 72, 353–363.

Stuart-Fox, D M, Firth, D, Moussalli, A and Whiting, M J (2006). Multiple signals in chameleon contests: designing and analysing animal contests as a tournament. Animal Behaviour 71, 1263–1271.

Firth, D. and Menezes, R. X. de (2004). Quasi-variances. Biometrika 91, 65–80.

Firth, D. (2003). Overcoming the reference category problem in the presentation of statistical models. Sociological Methodology 33, 1–18.

Election exit polls
David's research work is behind the accurate exit-poll predictions that have been made at the last six UK general elections (in 2001, 2005, 2010, 2015, 2017 and 2019). For more details see exit polling explained.
Far-reaching statistical theory!
What connects all these: nuclear weapons, rattlesnakes, coral, pneumonia, cancer, basketball?

Contact details:

Email: d dot firth at warwick dot ac dot uk
PGP public key: df-key.txt (fingerprint 219F 8A60 C58F 2057 6ACF 715C A534 271C B684 1FAC )

Postal address:
  Professor D Firth
Dept of Statistics
University of Warwick
Coventry, CV4 7AL
United Kingdom

Office location: MSB 4.08, new Mathematical Sciences building.

Zeeman building in October

Mastodon:
@firthstat@fediscience.org

Twitter:
@firthstat (no longer in use)